- Hintikka holds that Descartes' famous "Cogito, ergo sum" is not deductively valid. He says: "My point may perhaps be illustrated by means of an example constructed for us by Shakespeare. Hamlet did think a great many things; does it follow that he existed?" But "Descartes realized, however dimly, the existential inconsistency of the sentence 'I don't exist' and therefore the existential self-verifiability of 'I exist."' Existential inconsistency or self-verifiability is of ". . . performatory character. It depends on ... a certain person's act of uttering a sentence." Sentences such as 'I exist' or 'I am' "verify themselves when they are expressly uttered or otherwise professed." Thus the Cogito is, as an argument, not deductively valid, but is, as a performance, existentially self-verifying.
- In the reviewer's opinion, if not existing is no obstacle to thinking, then it should not stand in the way of saying "I exist" either. If 'Hamlet thought' is true, then it seems that "Hamlet said 'I exist"' only misses being true by virtue of the historical accident that Shakespeare did not give his character this particular line. If Hamlet had been given the line 'I exist' would Hamlet have existed?
- This absurdity could be avoided by adding the qualification that 'I exist' is only "existentially self-verifying" if the creature that utters it exists. But then one might as well say that 'I walk' is "existentially self-verifying" if the creature that utters it walks. After all, it is not the saying of 'I exist' that makes it true – it is the existing, just as it is the walking, and not the saying, that makes 'I walk' true.
- Hintikka says that "systems of logic in which (1) B(a) ⊃ (∃x)(x = a) can be proved ... are based on important existential presuppositions ... they make more or less tacit use of the assumption that all singular terms with which we have to deal really refer to (designate) some actually existing individual ... this amounts to assuming that the term which replaces a in (1) must not be empty."
- The reviewer believes that the semantics2 for systems of logic in which (1) is provable need not be based on replacing the individual constants with terms from some other language. The constants may simply be made to stand for individuals. And the provability of (1) does not entail a preference for actually or presently (or anything besides consistently) existing individuals. Any nameable individuals will do. It may be argued that there are no individuals but actually existing individuals, but the logician who accepts (1) need not worry about the outcome of this argument. If the term 'Hamlet' in 'Hamlet thought' designates an individual, then that individual can be designated by an individual constant of the formal system. If 'Hamlet' does not designate an individual, then the meaning of 'Hamlet thought' cannot be captured by an atomic formula of the system. (Whether its meaning can be captured by an atomic formula in any case is another matter, which does not essentially involve the question of whether (1) should be provable.)
- Furthermore, while "There is such a person as Hamlet" is false, "There is such a literary character as Hamlet" is true. (This latter could be reconstructed in some way with an eye to ontological economy, but so could "Hamlet thought.") In view of this, the question whether the term 'Hamlet' is "empty" is very obscure.
Review of "Hintikka (Jaakko) - Cogito, Ergo Sum: Inference or Performance?".
Footnote 1: Omitting the first two paragraphs, which review "Lemmon (E.J.) - On Sentences Verifiable by Their Use", and have been placed there as that paper’s abstract.
Text Colour Conventions (see disclaimer)
- Blue: Text by me; © Theo Todman, 2020
- Mauve: Text by correspondent(s) or other author(s); © the author(s)